LABEL FREE MOLECULAR IMAGING OF TUMOUR SECTIONS FOR TWO AND THREE DIMENSIONAL TISSUE CLASSIFICATION AND PATHWAY MAPPING

Significance of the topic

Mass spectrometry imaging enables direct, label-free mapping of metabolites and lipids in tissue sections, offering insights into tumour heterogeneity and metabolic pathways without complex sample preparation.

Aims and Study Overview

This study applies desorption electrospray ionisation (DESI) mass spectrometry imaging to classify tumour regions based on molecular fingerprints and to reconstruct two- and three-dimensional metabolic maps in cancer models.

Methodology and Instrumentation

• Instrument: Xevo G2-XS Q-TOF mass spectrometer with 2D DESI stage (Waters, UK; Prosolia, USA).
• Sample preparation: Snap-frozen tissue sections (20 µm) mounted on glass slides, stored at –80 °C.
• Data analysis: High Definition Imaging 1.4, SCiLS Lab for statistical modeling, MATLAB for rendering.
• Biological models: Invasive ductal carcinoma sections, U87 glioma xenografts, HCT-116 spheroids in agar-coated plates.

Main Results and Discussion

DESI-MS captured endogenous metabolites (e.g., lactate, pyruvate) and phospholipids in a single experiment. Co-registration with H&E staining enabled region-of-interest selection, ROC analysis identified discriminant ions, and machine learning models classified tissue pixels automatically. Serial section acquisition supported 3D chemical reconstructions of tumour spheroids and xenografts.

Benefits and Practical Applications

• Label-free analysis preserves native molecular distributions and simplifies workflows.
• Simultaneous detection of low- and high-molecular-weight species yields comprehensive metabolic profiles.
• Automated tissue classification enhances histopathology and biomarker discovery.
• 3D reconstructions align ex vivo molecular data with in vivo imaging modalities for deeper insights.

Future Trends and Potential Applications

Developments in semi-quantitative and absolute quantitation methods will refine metabolic flux analysis. Expanding molecular coverage and optimizing machine learning pipelines may improve diagnostic accuracy. Increased automation and throughput will enable large-scale studies, drug distribution mapping, and integration with multimodal imaging.

Conclusion

DESI-MS imaging offers a powerful, label-free approach for molecular classification and metabolic mapping of tumour tissues in two and three dimensions. Ongoing technological and computational advances are poised to enhance quantitation, expand molecular coverage, and streamline workflows for both research and clinical applications.

References

1. Swales JG et al. Analytical Chemistry. 2014;86:8473–8480.
2. Miura D et al. Analytical Chemistry. 2010;82:9789–9796.
3. McDonnell LA et al. Journal of Proteomics. 2010;73:1921–1944.
4. Dekker TJA et al. Journal of Proteome Research. 2014;13:4730–4738.
5. Palmer AD, Alexandrov T. Analytical Chemistry. 2015;87:4055–4062.